© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 35760
    Prev
    Next
    INDEX
    Explanations

    <pickle> </pickle> tags near code or data serialization

    np_acts-logits-general · gemini-2.5-flash-lite

    This neuron spots markup‐tag delimiters (the “<” and corresponding tag names like “pickle,” “meta,” etc.) in XML/HTML‐style text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     their
    -1.45
     its
    -1.33
    $.
    -1.05
     musica
    -1.04
     has
    -1.03
     it
    -1.02
    честву
    -1.00
     regia
    -1.00
     them
    -0.99
    Ὀ
    -0.98
    POSITIVE LOGITS
    Descriere
    1.30
    Masukkan
    1.28
     также
    1.23
    moteur
    1.20
    Pilih
    1.16
     seleccionar
    1.15
    alians
    1.13
     partecipare
    1.13
    secara
    1.11
    dengan
    1.10
    Activations Density 0.001%

    No Known Activations